|
(11) | EP 3 001 670 A2 |
| (12) | EUROPEAN PATENT APPLICATION |
|
|
|
|
|||||||||||||||||||||||
|
||||||||||||||||||||||||
| (54) | EFFICIENT DARK CURRENT SUBTRACTION IN AN IMAGE SENSOR |
| (57) A method for dark current subtraction which enables a dark frame to be reused for
dark current subtraction for multiple image frames. The dark frame is reused by scaling
it according to changes in the dark current levels associated with the dark frame
and the image frames.
|
BACKGROUND OF THE INVENTION
Field of Invention
[0001] The present invention pertains to the field of image sensors. More particularly, this invention relates to efficient dark current subtraction for an image sensor.
Art Background
[0002] A typical image sensor includes a set of sensing elements which generate electrical charge in response to light. For example, one type of image sensor is a charge-coupled device (CCD) array. A typical CCD array includes an array of CCD sensing elements and circuitry for sampling the charge levels of the CCD sensing elements.
[0003] A variety of images sensors including CCD arrays have light sensing elements which accumulate an electrical charge even in the absence of light. The electrical charge that accumulates in a sensing element in the absence of light is commonly referred to as dark current. Typically, the dark current in an image sensor creates an undesirable "dark" image that overlays the optical image.
[0004] One prior method for removing the effects of dark current in an image sensor is to perform dark current subtraction. Typically, an image is obtained from the image sensor when it is illuminated by light from an image scene and an image is obtained from the image sensor when it is not illuminated. The image obtained when the image sensor is illuminated may be referred to as an image frame. The image obtained when the image sensor is not illuminated may be referred to as a dark frame. The dark frame is usually subtracted from the image frame to yield an image frame without the dark current component.
[0005] Prior methods for dark current subtraction usually obtain a dark frame each time an image frame is obtained because the dark current typically varies according to temperature and exposure. Unfortunately, the time consumed in obtaining a dark frame each time an image frame is obtained usually slows the rate at which useful images may be obtained with an image sensor.
SUMMARY OF THE INVENTION
[0006] A method for dark current subtraction is disclosed which enables a dark frame to be reused for dark current subtraction for multiple image frames. The dark frame is reused by scaling it according to changes in the dark current levels associated with the dark frame and the image frames. In one embodiment, the changes in dark current levels are determined by examining changes in charge samples from optically black sensing elements and dummy samples from circuitry in the image sensor.
[0007] Other features and advantages of the present invention will be apparent from the detailed description that follows.
[0008] Certain aspects of the invention are discussed in the following clauses.
Clause 1: There may be provided a method for dark current subtraction, comprising the steps of: obtaining a dark frame from an image sensor; reusing the dark frame for dark current subtraction by scaling the dark frame in response to changes in dark current levels in the image sensor (18).
Clause 2: There may be provided a method according to clause 1, wherein the step of reusing the dark frame for dark current subtraction comprises the steps of: obtaining an image frame from the image sensor (18); determining a difference between dark current levels associated with the image frame and dark current levels associated with the dark frame; determining a scaled dark frame from the dark frame according to the difference; subtracting the scaled dark frame from the image frame.
Clause 3: There may be provided a method according to clause 2, wherein the step of obtaining a dark frame comprises the steps of: obtaining a dark sample from each of a set of active elements (32) of the image sensor (18); obtaining a dark sample from each of a set of optically black elements (30, 34) of the image sensor (18); obtaining a set of dummy samples that represent a low charge level in the image sensor (18).
Clause 4: There may be provided a method according to clause 1, wherein the step of obtaining a set of dummy samples comprises the step of over-clocking an output register in the image sensor (18).
Clause 5: There may be provided a method according to clause 3, wherein the step of obtaining an image frame comprises the steps of: obtaining an image sample from each active element; obtaining an image sample from each optically black element; obtaining a set of dummy samples from the image sensor (18).
Clause 6: There may be provided a method according to clause 5, wherein the step of determining a difference between the dark current levels comprises the steps of: determining a difference between a subset of the image samples from the optically black elements (30, 34) and a subset of the dummy samples; determining a difference between a subset of the dark samples from the optically black elements (30, 34) and a subset of the dummy samples.
Clause 7: There may be provided a method according to clause 6, wherein the step of determining a difference between a subset of the image samples from the optically black elements (30, 34) and a subset of the dummy samples comprises the steps of: determining an average of the subset of the image samples from the optically black elements (30, 34); determining an average of the subset of the dummy samples; determining a difference between the averages.
Clause 8: There may be provided a method according to clause 6, wherein the step of determining a difference between a subset of the dark samples from the optically black elements (30, 34) and a subset of the dummy samples comprises the steps of: determining an average of the subset of the dark samples from the optically black elements (30, 34); determining an average of the subset of the dummy samples; determining a difference between the averages.
Clause 9: There may be provided a method according to clause 6, wherein the step of determining a scaled dark frame comprises the step of multiplying the dark frame by a ratio of the difference between the subsets.
Clause 10: There may be provided an apparatus for dark current subtraction, comprising: means for obtaining a dark frame from an image sensor (18); means for reusing the dark frame for dark current subtraction by scaling the dark frame in response to changes in dark current levels in the image sensor (18).
Clause 11: There may be provided an apparatus according to clause 10, wherein the means for reusing the dark frame for dark current subtraction comprises: means for obtaining an image frame from the image sensor (18); means for determining a difference between dark current levels associated with the image frame and dark current levels associated with the dark frame; means for determining a scaled dark frame from the dark frame according to the difference; means for subtracting the scaled dark frame from the image frame.
Clause 12: There may be provided an apparatus according to clause 11, wherein the
means for obtaining a dark frame comprises : means for obtaining a dark sample from
each of a set of active elements (32) of the image sensor (18); means for obtaining
a dark sample from each of a set of optically black elements (30, 34) of the image
sensor (18);
means for obtaining a set of dummy samples that represent a low charge level in the
image sensor (18).
Clause 13: There may be provided an apparatus according to clause 10, wherein the means for obtaining a set of dummy samples comprises means for over-clocking an output register in the image sensor (18).
Clause 14: There may be provided an apparatus according to clause 12, wherein the means for obtaining an image frame comprises :
means for obtaining an image sample from each active element;
means for obtaining an image sample from each optically black element;
means for obtaining a set of dummy samples from the image sensor (18).
Clause 15: There may be provided an apparatus according to clause 14, wherein the means for determining a difference between the dark current levels comprises: means for determining a difference between a subset of the image samples from the optically black elements (30, 34) and a subset of the dummy samples; means for determining a difference between a subset of the dark samples from the optically black elements (30, 34) and a subset of the dummy samples.
Clause 16: There may be provided an apparatus according to clause 15, wherein the
means for determining a difference between a subset of the image samples from the
optically black elements (30, 34) and a subset of the dummy samples comprises: means
for determining an average of the subset of the image samples from the optically black
elements (30, 34); means for determining an average of the subset of the dummy samples;
means for determining a difference between the averages.
Clause 17: There may be provided an apparatus according to clause 15, wherein the means for determining a difference between a subset of the dark samples from the optically black elements (30, 34) and a subset of the dummy samples comprises: means for determining an average of the subset of the dark samples from the optically black elements (30, 34); means for determining an average of the subset of the dummy samples; means for determining a difference between the averages.
Clause 18: There may be provided an apparatus according to clause 15, wherein the
means for determining a scaled dark frame comprises means for multiplying the dark
frame by a ratio of the
difference between the subsets.
[0009] In one particular aspect, the invention concerns an apparatus for dark current subtraction, comprising an image sensor (18) including an array with a set of active elements (32) and a set of optically black elements (30, 34);
wherein the apparatus is adapted for obtaining a dark frame (100) and an image frame (102) from the image sensor (18), each frame including a sample from each of a set of active elements (32) and a sample from each of a set of optically black elements (30, 34) in the image sensor (18);
wherein the dark frame is reuseable for dark current subtraction for multiple image frames obtained from the image sensor (18); and
wherein the apparatus further comprises an output register (50) configured for reading out the samples (ACTIVE, OB) from the active elements (32) and the optically black elements (30, 34);
wherein the dark frame and the image frame each additionally include a set of dummy samples (DUMMY) that represent a low charge level in the image sensor (18); and
wherein the apparatus further comprises an image processor (22), configured for clocking out rows of samples serially from the output register (50) by generating transitions or edges of a clock signal portion of control signals (26) to shift out the samples onto an output signal (28).
[0010] In a further particular aspect, the invention concerns an apparatus for dark current subtraction, comprising an image sensor (18) including an array with a set of active elements (32) and a set of optically black elements (30, 34);
wherein the apparatus is adapted for obtaining a dark frame (100) and an image frame (102) from the image sensor (18), each frame including a sample from each of a set of active elements (32) and a sample from each of a set of optically black elements (30, 34) in the image sensor (18);
wherein the dark frame is reuseable by scaling according to a difference between dark current levels associated with the image frame and dark current levels associated with the dark frame, for dark current subtraction for multiple image frames obtained from the image sensor (18);
wherein the dark frame and the image frame each additionally include a set of dummy samples (DUMMY) from circuitry in the image sensor, that represent a low charge level in the image sensor (18); said difference between dark current levels associated with the image frame and dark current levels associated with the dark frame being
determinable from the samples from the optically black elements and the dummy samples.
[0011] In still a further aspect of the invention, there is provided a digital camera comprising any of the apparatuses discussed above, or adapted to be used in accordance with any of the methods above.
BRIEF DESCRIPTION OF THE DRAWINGS
[0012] The present invention is described with respect to particular exemplary embodiments thereof and reference is accordingly made to the drawings in which:
Figure 1 shows a method for dark current subtraction according to the present teachings;
Figure 2 shows a camera which incorporates the present teachings;
Figure 3 shows an example arrangement of elements in an image sensor including active elements and optically black elements and a representation of dummy elements;
Figure 4 shows a data frame obtained via an output signal of an image sensor.
DETAILED DESCRIPTION
[0013] Figure 1 shows a method for dark current subtraction according to the present teachings. This method may be used, for example to perform dark current subtraction for an image sensor in a digital still camera or digital video camera.
[0014] At step 100, a dark frame is obtained from the image sensor. Step 100 may be performed, for example, by closing a shutter on the camera that contains the image sensor and obtaining samples from the image sensor.
[0015] At step 102, the dark frame obtained at step 100 is reused for dark current subtraction for multiple image frames obtained from the image sensor. The dark frame obtained at step 100 may be reused at step 102 by obtaining an image frame from the image sensor when it is illuminated by light from an image scene and then determining a difference between dark current levels associated with the image frame obtained at step 102 and dark current levels associated with the dark frame obtained at step 100. The dark frame obtained at step 100 may then be scaled according to the difference and then subtracted from the image frame obtained at step 102.
[0016] The dark frame obtained at step 100 and the image frame obtained at step 102 each include a sample from each of a set of active elements of the image sensor and a sample from each of a set of optically black elements in the image sensor. In addition, the dark frame obtained at step 100 and the image frame obtained at step 102 each include a set of dummy samples that represent a low charge level in the image sensor. In one embodiment, the dummy samples are obtained by over-clocking an output register in the image sensor.
[0017] In one embodiment, the difference between the dark current levels associated with the dark frame and the image frame is determined by determining differences between samples from the optically black elements the dummy samples.
[0018] Figure 2 shows a camera 10 which incorporates the present teachings. The camera 10 includes a lens mechanism 12 that conducts light from an image scene through an aperture mechanism 14 and a shutter mechanism 16 to an image sensor 18. An image processor 22 controls the aperture mechanism 14 and the shutter mechanism 16 via a set of control signals 24.
[0019] The image sensor 18 accumulates electrical
charge in its sensing elements during exposure periods which the image processor 22 controls via a set of control signals 26. The image sensor 18 also accumulates charge in its sensing elements during exposure periods as a result of dark currents in the image sensor 18.
[0020] The image processor 22 generates signals via the control signals 24 and 26 to start exposure periods, stop exposure periods and to read out samples from the image sensor 18. The samples obtained from the image sensor 18 are provided to an analog-to-digital (A/D) converter 20 via an output signal 28. The A/D converter 20 digitizes the samples carried by the output signal 28 and provides the corresponding digitized samples to the image processor 22.
[0021] The image processor 22 controls the aperture mechanism 14 and the shutter mechanism 16, and the image sensor 18 to obtain digitized samples for a dark frame at step 100 and for an image frame at step 102. For example, a dark frame may be obtained by closing the shutter mechanism 16 using the control signals 24 and then using the control signals 26 to obtain samples from the image sensor 18. An image frame may be obtained by setting an exposure with the aperture mechanism 14 and the shutter mechanism 16 using the control signals 24 and then using the control signals 26 to obtain samples from the image sensor 18.
[0022] In one embodiment, the camera 10 is a still image camera. In another embodiment, the camera 10 is a video camera. The image sensor 18 may be a CCD array or complementary metal-oxide semiconductor (CMOS) array. The image processor 22 including dark current subtraction functionality may be implemented in hardware and/or software or firmware. The aperture mechanism 14 and the shutter mechanism 16 may be any known mechanisms useful in cameras.
[0023] Figure 3 shows an example arrangement of elements in the image sensor 18. In this example, the image sensor 18 includes an array with a set of active elements 32, two sets of optically black elements 30 and 34, and an output register 50. In one embodiment, the active elements 32 are CCD light sensing elements and the optically black elements 30 an 34 are CCD light sensing elements that are covered with a material that blocks light.
[0024] The output register 50 is used to read out charge samples from the active elements 32 and the optically black elements 30 and 34. Circuitry (not shown) in the image sensor 18 selects the rows the active elements 32 and the optically black elements 30 and 34 in sequence. Charge from a selected row of the active elements 32 is provided to the output register 50 via a set of signal lines 42 and charge from a selected row of the optically black elements 30 and 34 is provided to the output register 50 via sets of signal lines 40 and 44.
[0025] Once a row of charge samples has been collected in the output register 50, the image processor 22 clocks out the row of samples serially onto the output signal 28 using the control signals 26. In this example, a row of samples includes 2 optically black samples, 8 active samples and 2 optically black samples. The image processor 22 obtains these samples by generating 12 transitions or edges of clock signal portion of the control signals 26 to shift out these samples onto the output signal 28.
[0026] The image processor 22 obtains dummy samples from the image sensor 18 by over-clocking the output register 50. The dummy samples represent a low charge level in the image sensor 18 for purposes of scaling dark current when reusing a dark frame. The dummy samples may be treated as if the image sensor 18 included an area 36 of dummy elements which are devoid of dark. For example, the image processor 22 obtains 4 dummy samples for a row by generating 4 extra transitions or edges of the clock signal on the control signals 26 after clocking out the samples from the optically black and active elements.
[0027] Figure 4 shows a data frame 60 obtained from the image sensor 18 via the output signal 28. The data frame 60 may be a dark frame obtained at step 100 or an image frame obtained at step 102. The data frame 60 includes a series of row samples (row 1-n) wherein n is the number of rows of sensing elements in the image sensor 18.
[0028] Each row 1-n includes a set of samples (OB) from the optically black elements 30 followed by a set of samples (ACTIVE) from the active elements 32 followed by a set of samples (OB) from the optically black elements 34 followed by a set of dummy samples (DUMMY) from what is represented as the area 36.
[0029] The image processor 22 determines an average intensity of the samples from the optically black elements in a dark frame (DOB) and an average intensity of the dummy samples in the dark frame (DDUMMY). The averages DOB and DDUMMY may be computed for all of the samples or a subset of the samples in the dark frame.
[0030] The image processor 22 determines an average intensity of the samples from the optically black elements in an image frame (IOB) and an average intensity of the dummy samples in the image frame (IDUMMY). The averages IOB and IDUMMY may be computed for all of the samples or a subset of the samples in the image frame.
[0031] The samples from the active elements 32 in a dark frame obtained at step 100 are DACTIVE (1, 1) through DACTIVE (n,m) where n is the number of rows and m is the number of columns in the active elements 32. The samples from the active elements 32 for an image frame obtained at step 102 are IACTIVE (1,1) through IACTIVE (n, m). The dark frame obtained at step 100 is scaled at step 102 in one embodiment by multiplying DACTIVE (1 , 1) through DACTIVE (n,m) by the following scale factor:
[0032] The scaled dark frame obtained using the above scale factor is subtracted from IACTIVE (1, 1) through IACTIVE (n,m) element by element when performing dark current subtraction at step 102.
[0033] The time taken to generate a scaled dark frame from a stored dark frame using the above technique is less than the time taken to close the shutter mechanism 16 and obtain a new dark frame. This reuse of a dark frame enables an increased rate of image frame sampling in the camera 10. A dark frame may be stored in a memory associated with the image processor 22 and re-scaled and reused for each acquired image frame thereafter.
means for obtaining a dark frame from an image sensor (18);
means for reusing the dark frame for dark current subtraction by scaling the dark frame in response to changes in dark current levels in the image sensor (18).
wherein the apparatus is adapted for obtaining a dark frame (100) and an image frame (102) from the image sensor (18), each frame including a sample from each of a set of active elements (32) and a sample from each of a set of optically black elements (30, 34) in the image sensor (18);
wherein the dark frame is reuseable for dark current subtraction for multiple image frames obtained from the image sensor (18); and
wherein the apparatus further comprises an output register (50) configured for reading out the samples (ACTIVE, OB) from the active elements (32) and the optically black elements (30, 34);
wherein
the dark frame and the image frame each additionally include a set of dummy samples (DUMMY) that represent a low charge level in the image sensor (18); and
wherein the apparatus further comprises an image processor (22), configured for clocking out rows of samples serially from the output register (50) by generating transitions or edges of a clock signal portion of control signals (26) to shift out the samples onto an output signal (28).
wherein the apparatus is adapted for obtaining a dark frame (100) and an image frame (102) from the image sensor (18), each frame including a sample from each of a set of active elements (32) and a sample from each of a set of optically black elements (30, 34) in the image sensor (18);
wherein the dark frame is reuseable by scaling according to a difference between dark current levels associated with the image frame and dark current levels associated with the dark frame, for dark current subtraction for multiple image frames obtained from the image sensor (18);
wherein, the dark frame and the image frame each additionally include a set of dummy samples (DUMMY) from circuitry in the image sensor, that represent a low charge level in the image sensor (18); said difference between dark current levels associated with the image frame and dark current levels associated with the dark frame being determinable from the samples from the optically black elements and the dummy samples.
the image sensor comprises means for obtaining a set of dummy samples that represent a low charge level in the image sensor (18) such that the dummy samples (DUMMY) are treated as an area of dummy elements included on the image sensor
or
an output register in the image sensor is arranged to obtain dummy samples by over-clocking.
- obtaining a dark sample from each of a set of active elements (32) of the image sensor (18);
- obtaining a dark sample from each of a set of optically black elements (30, 34) of the image sensor (18);
- obtaining a set of dummy samples that represent a low charge level in the image sensor (18).
- obtaining an image sample from each active element;
- obtaining an image sample from each optically black element;
- obtaining a set of dummy samples from the image sensor (18).
- determining a difference between a subset of the image samples from the optically black elements (30, 34) and a subset of the dummy samples;
- determining a difference between a subset of the dark samples from the optically
black elements (30, 34) and a subset of the dummy samples,
wherein preferably the step of determining a difference between a subset of the image
samples from the optically black elements (30, 34) and a subset of the dummy samples
comprises the steps of:
- determining an average of the subset of the image samples from the optically black elements (30, 34);
- determining an average of the subset of the dummy samples;
- determining a difference between the averages,
wherein preferably the step of determining a difference between a subset of the dark samples from the optically black elements (30, 34) and a subset of the dummy samples comprises the steps of:- determining an average of the subset of the dark samples from the optically black elements (30, 34);
- determining an average of the subset of the dummy samples;
- determining a difference between the averages,
wherein preferably the step of determining a scaled dark frame comprises the step of multiplying the dark frame by a ratio of the difference between the subsets.- an image sensor (18), comprising sensing elements for obtaining a dark frame and a subsequent image frame, wherein the sensing elements include:
∘ active elements (32), for acquiring active image samples during acquisition of the dark frame and of the image frame;
∘ optically black elements (30, 34), for acquiring optically black samples during acquisition of the dark frame and of the image frame;
∘ means for acquiring dummy samples that represent a low charge level in the image sensor;
- an image processor (22), configured for:
o controlling the sensing elements during acquisition of the dark frame and of the image frame, and for
o reusing the dark frame for dark current subtraction, in response to changes in dark current levels associated with the image frame, with respect to dark current levels associated with the dark frame, determinable from the samples from the optically black elements and the dummy samples.
* scaling the active image samples of the dark frame based on the optically black samples and the dummy samples;
* subtracting the scaled active image samples of the dark frame from the active image samples of the image frame.
- means for obtaining an image frame from the image sensor (18);
- means for determining a difference between dark current levels associated with the image frame and dark current levels associated with the dark frame;
- means for determining a scaled dark frame from the dark frame according to the difference;
- means for subtracting the scaled dark frame from the image frame.
- means for obtaining a dark sample from each of a set of active elements (32) of the image sensor (18);
- means for obtaining a dark sample from each of a set of optically black elements (30, 34) of the image sensor (18);
- means for obtaining a set of dummy samples that represent a low charge level in the image sensor (18).
and/or
wherein the means for determining a difference between a subset of the dark samples from the optically black elements (30, 34) and a subset of the dummy samples comprises: means for determining an average of the subset of the dark samples from the optically black elements (30, 34); means for determining an average of the subset of the dummy samples; means for determining a difference between the averages.